Forecasting indoor environment, especially solving contaminant transport problem in a dynamic environment, is a difficult task since the physical states of the building environment could change rapidly over time. By using conventional methods to predict future indoor air contaminants, the uncertainties of the modeled environment will cause the predicted physical states to depart from reality as the model evolves over time. In this project, a new application of Ensemble Kalman Filter (EnKF) to forecast indoor environment is presented by using a mass balance model in OpenDA while the contaminant concentration can be predicted without using local measurements. The results indicate that by using EnKF, the predictability of the indoor air model is improved significantly. More information about this project and the research group can be found at the project website: Link to the paper: